mlsanityData to 13 Oct 2026

SEA-LION-v4.8: A Technical Report

arXiv 2609.18310 ↗First posted 16 Sep 2026cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

We introduce Nemotron-SEA-LION-v4.8, a family of Southeast Asian Languages In One Network (SEA-LION) models built upon NVIDIA Nemotron 3. The family includes 30B-A3B and 120B-A12B models, with both continued-pretrained base checkpoints and post-trained variants. We adapt the models using Southeast Asian, reasoning, code, and multilingual parallel data, followed by post-training with supervised fine-tuning and online on-policy distillation. On SEA-HELM, the 30B-A3B model improves the overall SEA score from 46.06 to 51.57, while the 120B-A12B model improves from 49.30 to 63.44. Across seven Southeast Asian languages, we observe broad capability gains with the 120B-A12B model showing broader and more consistent improvements across tasks.

Key facts. No citing papers in the map yet (data through 13 Oct 2026). Builds on 11 of its 14 references that are in the map. First posted on arXiv 16 Sep 2026.

Builds on

11 of 14 references.

References · 11

  1. 2024Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual Evaluationcited by 182
  2. 2023Continual Pre-Training of Large Language Models: How to (re)warm your model?cited by 137
  3. 2024Instruction-tuned Language Models are Better Knowledge Learnerscited by 55
  4. 2023Typhoon: Thai Large Language Modelscited by 29
  5. 2025Front-Loading Reasoning: The Synergy between Pretraining and Post-Training Datacited by 26
  6. 2025SEA-HELM: Southeast Asian Holistic Evaluation of Language Modelscited by 18
  7. 2023BHASA: A Holistic Southeast Asian Linguistic and Cultural Evaluation Suite for Large Language Modelscited by 16
  8. 2025How Learning Rate Decay Wastes Your Best Data in Curriculum-Based LLM Pretrainingcited by 13
  9. 2026Pre-training LLM without Learning Rate Decay Enhances Supervised Fine-Tuningcited by 13
  10. 2026OpenSeal: Good, Fast, and Cheap Construction of an Open-Source Southeast Asian LLM via Parallel Datacited by 3
  11. 2026BURMESE-SAN: Burmese NLP Benchmark for Evaluating Large Language Modelscited by 1

Built on by

Not cited yet.